vccep pilot: progress on evaluating children’s risks … et al accepted.pdf · vccep pilot:...
TRANSCRIPT
1
VCCEP Pilot: Progress on Evaluating Children’s Risks and Data Needs
Submitted to:
Risk Analysis
Original: November 11, 2005 Revised: January 11, 2005 Accepted for Publication
Authors:
Pamela R. D. Williams Jacqueline Patterson
Daniel W. Briggs
2
ABSTRACT
The Voluntary Children’s Chemical Evaluation Program (VCCEP) is designed to provide
information to the public on children’s potential health risks associated with chemical exposures.
The key question of the VCCEP is whether the potential hazards, exposures, and risks to children
have been adequately characterized, and, if not, what additional data are necessary. To answer
this question, manufacturers or importers of 23 chemicals were asked by the U.S. EPA to
sponsor their chemicals in the first tier of a pilot program. These chemicals were selected for
evaluation because they have been found as contaminants in human tissue or fluids (adipose
tissue, blood, breath, breast milk, or urine); food and water children may eat and drink; or air
children may breathe (including residential or school air). Under the VCCEP framework,
sponsoring companies agree to prepare Tier 1 hazard, exposure, and risk assessments on the
individual chemicals, and identify the need for additional data. These assessment documents are
submitted to the U.S. EPA and subsequently undergo review by experts in an independent peer
consultation meeting that is open to the public. Following this peer consultation process, the
U.S. EPA reviews each submission and makes a data needs determination, which may include
requesting further data collection or generation by the sponsor. Sponsoring companies then
decide whether to volunteer for the next tier and collect or generate the requested data.
The purpose of this paper is to describe the process and to review and present the key findings
from the first set of chemicals that have been fully or partially evaluated under the pilot program
(vinylidene chloride, decabromodiphenyl ether, pentabromodiphenyl ether, octabromodiphenyl
ether, acetone, methyl ethyl ketone, decane, undecane, and dodecane). Specifically, we provide
a brief summary of the sponsors’ submissions, the peer consultation panels’ discussions, and the
U.S. EPA’s data needs decisions. Although we do not attempt to conduct independent analyses
3
of the underlying data, we do identify a number of common themes that have emerged during
implementation of the pilot program and discuss several key issues that could become important
in the future. The information presented here should be useful for various parties interested in
the progress of the VCCEP and the results of the initial (Tier 1) children’s assessments.
KEY WORDS VCCEP, children’s health, risk assessment, industrial chemicals
4
INTRODUCTION
Children’s health issues have become an increasingly important topic of scientific research
over the last decade. In particular, greater attention has focused on the potential for children’s
increased (or reduced) susceptibility to chemicals and on children’s unique behavior and
exposure patterns (ATSDR 1997; Armstrong et al. 2000; Bruckner 2000; Calabrese 2001;
Charnley 2001; Charnley and Putzrah 2001; Cohen-Hubal et al. 2000; Dourson et al. 2002; Etzel
et al. 1999; Juberg 2002; Scheuplein et al. 2002; Schneider and Freeman 2000; U.S. EPA 2003;
Weaver et al. 1998). Approaches for conducting child-specific exposure and risk analyses have
also populated the literature in recent years (AIHC 1994; Armstrong et al. 2000; Finley et al.
1994; Cohen-Hubal et al. 2000; U.S. EPA 1989, 1997, 2002b; Resiss et al. 2003; Williams et al.
2003). In response to growing concerns about children’s health issues in the United States,
several federal laws and a number of nationwide policies and initiatives aimed at studying and
protecting children from environmental health hazards have been enacted. These include the
Food Quality Protection Act, the Children’s Health Act, the U.S. Environmental Protection
Agency’s (U.S. EPA) Office of Children’s Health Protection and other Agency initiatives, and
various programs initiated by the Food and Drug Administration and Centers for Disease Control
and Prevention (CDC 2001, 2003; CHA 2000; FDA 2001; FQPA 1996; U.S. EPA 2000a,
2002a).
The Voluntary Children’s Chemical Evaluation Program (VCCEP), which is part of the U.S.
EPA’s Chemical-Right-to-Know Initiative, represents one of several child-centered programs
started in 2000. This unique pilot program was designed to provide data to the public that would
increase their understanding of potential health risks to children from chemical exposures and to
encourage the participation of interested stakeholders (U.S. EPA 2000b). The VCCEP is based
5
on the voluntary participation of chemical manufacturers and importers. The basic structure of
the program is that the available hazard and exposure data for a chemical are evaluated and used
to characterize risks to children and prospective parents. The identification of additional data
needs is informed by the findings of the health risk assessment. This consideration of the
available exposure and hazard data to inform the need for analyses or data collection represents a
paradigm shift from previous testing programs.
In the VCCEP pilot, the U.S. EPA asked the manufacturers and importers of 23 chemicals to
volunteer to sponsor their chemicals for evaluation of risks to children’s health. These chemicals
were selected by the U.S. EPA because they have been found as contaminants in human tissue or
fluids (e.g., adipose tissue, blood breath, breast milk, urine); food and water children may eat and
drink; or air children may breathe (including residential or school air). Thirty-five companies
and ten consortia volunteered to sponsor 20 of the 23 chemicals (see Table 1). Under the
VCCEP framework, sponsoring companies agree to prepare Tier 1 hazard, exposure, and risk
assessments on the individual chemicals, and identify the need for additional data. The sponsors’
assessment documents are submitted to the U.S. EPA and subsequently undergo review by
experts in an independent peer consultation meeting that is open to the public. Following this
peer consultation process, the U.S. EPA reviews each submission and makes a data needs
determination, which may include requesting further data collection or generation by the
sponsor. Sponsoring companies then decide whether to volunteer for the next tier and collect or
generate the requested data.
Thus far, six chemicals have completed Tier 1 of VCCEP (vinylidene chloride,
decabromodiphenyl ether, pentabromodiphenyl ether, octabromodiphenyl ether, acetone, and
methyl ethyl ketone). Three additional chemicals have been partially completed under the pilot
6
program (decane, undecane, and dodecane). In this paper, we provide a brief summary of the
highlights of the pilot program, including the industry sponsors’ submissions, the peer
consultation panels’ discussions, and the U.S. EPA’s data needs decisions (where applicable).
We do not attempt to conduct independent analyses of the underlying data in the assessments,
rather we identify a number of common themes that have emerged during implementation of the
VCCEP and discuss several key issues that could be important in the future. The information
presented here should be useful for various parties interested in the progress of the VCCEP and
the results of the initial (Tier 1) children’s assessments.
INSERT TABLE 1 ABOUT HERE
BACKGROUND
The VCCEP pilot represents a novel approach for determining the need for collection of
further toxicity and exposure information for existing chemicals. The intent of the pilot is to
gain insight on how best to design and efficiently implement the ultimate program. The VCCEP
pilot is structured into three tiers to facilitate decisions on whether more testing or data are
needed to adequately characterize the risks to children or prospective parents from selected
chemicals. Sponsoring companies or consortia of manufacturers and importers of each chemical
volunteer for each tier separately, allowing them to choose whether to participate in subsequent
tiers if requested to do so by the U.S. EPA. By volunteering to be sponsors, the companies or
consortia agree to collect or develop health effects and exposure information about the
chemicals, to integrate this information into a risk assessment that characterizes the risks to
children, and to identify potential data needs.
The three tiers of the pilot program increase in scope and sophistication from basic
screening level assessments to more refined and robust analyses (see Figure 1). The tiered
7
assessments include four specific components: hazard assessment, exposure assessment, risk
assessment, and data needs. For the hazard assessment, the U.S. EPA has identified specific
toxicity tests that should be considered for each tier, ranging from acute and genetic toxicity
screening tests in Tier 1 to chronic, neurotoxicity, and developmental neurotoxicity studies in
Tier 3 (see Table 2). The purpose of this tiered approach was to reach a compromise among the
various stakeholders and to identify definite points to stop and assess what is known about each
chemical before deciding on the need for further testing or data compilation. While a list of
toxicity tests was provided for hazard, minimal guidance was provided on how to assess
children’s exposures. For example, the U.S. EPA (2000b) loosely defines a Tier 1 exposure
assessment as representing a “screening level” analysis in which conservative estimates are
based on readily available data, and a Tier 2 or 3 exposure assessment as more “advanced”
analyses that rely on well-designed monitoring studies or exposure models and include central
tendency and high end estimates. A 3-day workshop was held prior to the first VCCEP
submission to discuss various approaches for collecting and presenting exposure information
under the pilot program, but a standardized approach was not prescribed, allowing sponsors the
flexibility to develop their own frameworks (U.S. EPA 2002c). Although the VCCEP is
designed in tiers, with increasing complexity of data and analyses, sponsors are expected to use
all available data for the Tier 1 assessment.
To provide a wide-ranging scientific review of the sponsor’s assessments, each
submission must undergo an evaluation and discussion by a peer consultation panel comprised of
various independent experts. The peer consultation meetings are open to the public and all
interested parties are invited to submit technical comments on the submission for consideration
by the sponsors and the peer consultation panel members. The purpose of the peer consultation
8
is to provide a science-based evaluation of the data needs for each chemical based on the
assessment submitted by the sponsor, as well as, the expertise and knowledge of the panel. The
results of the peer consultation can assist the sponsor in identifying further evaluation and
research, if needed. A non-profit group, Toxicology Excellence for Risk Assessment (TERA),
has thus far conducted all of the VCCEP pilot peer consultation meetings using a transparent and
open process. Specifically, panelist candidates are nominated by various parties (including
VCCEP sponsors, the public, and TERA), and nominees are screened by TERA for potential
conflicts of interest and bias. The panel uses a “charge” or list of questions and issues to form
the structure for the panel discussions, which is designed to elicit panel members’ opinions about
key questions regarding the adequacy of available data and analyses to characterize risks to
children and prospective parents. The results of the panel discussions are summarized by TERA
(2005a) in a meeting report that is made available to the public via the web. Note that a peer
consultation does not seek panel consensus; rather, the individual opinions of each panel member
are noted in the meeting reports along with areas of agreement and disagreement among panel
members. Unlike the traditional peer review process, the peer consultations were designed to
elicit more active participation of the various stakeholders.
Following the peer consultation meeting for a chemical, the U.S. EPA reviews the sponsor’s
submission and the peer consultation panel’s report and determines whether the information for
each chemical is sufficient to adequately characterize the risks to children. If not, additional
testing or data are requested and a higher tiered analysis is recommended. This three-step
process (sponsor submission, panel review, U.S. EPA evaluation) is repeated for each tier. Note
that the sponsors of each chemical volunteer for one tier at a time, so that volunteering for Tier 1
does not obligate them to volunteer for Tier 2 or 3 if data from these subsequent tiers are deemed
9
necessary. According to the VCCEP framework (2000b), the risk information generated for each
chemical will ultimately be disseminated and communicated to the public, and any risks
identified as unacceptable will require mitigation.
It is noteworthy that the U.S. EPA conducted a series of stakeholder dialogs during the
development of the VCCEP, in which numerous concerns were raised about the structure and
implementation of the program. Specifically, individuals and groups representing industry,
environmental, public health, and/or animal welfare interests commented on the appropriateness
or validity of the chemicals selected, the tiered testing approach, the provision of renewed
commitments at each tier, industry’s preparation of assessments, and the peer consultation
process (U.S. EPA 2005a). The U.S. EPA therefore introduced the VCCEP as a pilot program,
so that it could gain insight into how best to design the program and to test the performance of
the peer consultation process.
INSERT FIGURE 1 ABOUT HERE
INSERT TABLE 2 ABOUT HERE
CHEMICALS EVALUATED UNDER THE VCCEP
To date, six chemicals have completed the VCCEP Tier 1 process, while three additional
chemicals have been partially completed (i.e., the U.S. EPA has not yet provided a data needs
decision for these latter chemicals). The following section provides a brief summary of the
current status and findings related to all nine chemicals. This includes the hazard, exposure, risk
assessment, and data needs assessments for each chemical as determined by the sponsors, peer
consultation panelists, and U.S. EPA. Because our review of each chemical is intended to
highlight only the most salient issues from the pilot program (and includes summary data
provided directly from the sponsors’ documents), it does not include all the findings or opinions
10
presented or expressed by the sponsors or individual panel members. More details about the
VCCEP documents are available from TERA (2005a) and the U.S. EPA (2005a).
Note that relatively early in the VCCEP pilot, the peer consultation panels adopted the term
“data gaps” in addition to “data needs” to represent two distinct concepts. Specifically, data gaps
were defined as areas for which information were not available or where there were significant
uncertainties, but which did not necessarily require additional research or data collection in order
to adequately characterize children’s health risks. For example, missing data on a particular
toxicity endpoint may not have a significant consequence on the results of the risk
characterization for a chemical if there is no or very low exposure to the chemical. With a very
large estimated margin of exposure, the collection of more data may not change the overall
conclusions of the risk assessment, and therefore the missing data on the toxicity endpoint may
be identified as a data gap. On the other hand, data needs were defined as data gaps for which
additional information would be required before the risks to children could be adequately
evaluated.
Vinylidene Chloride
The vinylidene chloride assessment prepared by the Dow Chemical Company was the
first industry submission under the VCCEP pilot (Dow 2002). Vinylidene chloride is used as a
chemical intermediate in the production of polymers and other chemicals. It can also be reacted
to produce polyvinylidene chloride latex and resin polymers. These production processes consist
primarily of closed-system operations in industrial settings. Primary applications include latex
for carpet backing, adhesives, and film coatings; flame retardant clothing; food packaging; and
water or oil resistant textiles.
11
The sponsor’s submission noted that vinylidene chloride has been extensively studied in a
variety of assays and test animals, and multiple studies are available for most Tier 1-3 categories.
Several existing hazard benchmarks were identified for this chemical, including the U.S. EPA
reference dose (RfD) of 50 µg/kg-day, a benchmark dose limit for liver toxicity (BMDL10) of
4600 µg/kg-day, the U.S. EPA reference concentration (RfC) of 200 µg/m3, and the California
reference exposure level (REL) of 70 µg/m3. Four exposure pathways were ultimately
considered in the quantitative analysis, including the inhalation of general ambient air, inhalation
of indoor air due to residual monomers in carpet backing, ingestion of contaminated water, and
ingestion of food from contact with packaging (dermal exposures were considered to be
insignificant or irrelevant). During the panel meeting, the sponsor also presented information on
potential inhalation of vapors migrating into indoor air of buildings located over contaminated
groundwater plumes (this pathway was not presented in the submission document because it is
not part of the “chain of commerce” for this chemical). Aggregate exposures based on the four
pathways were calculated using “typical” and “high end” estimates, and both a margin of safety
(MOS) and margin of exposure (MOE) approach were used to characterize chronic (non-cancer)
risks for children and prospective parents (see Table 3). The sponsor concluded that vinylidene
chloride did not pose a risk to children because exposures were likely to be inconsequential and
the hazard data do not suggest any unusual age-related sensitivity. The sponsor also thought the
available data were adequate to characterize risks to children.
INSERT TABLE 3 ABOUT HERE
A peer consultation panel reviewed the sponsor’s Tier 1 submission of vinylidene
chloride in January 2003 (TERA 2003a). Most panel members concluded that the hazard
benchmarks employed in the submission were appropriate, and some panelists thought the
12
toxicity studies covered essentially all the data sought from the three VCCEP testing tiers.
However, some panelists noted that developmental toxicity could not be ruled out because
reports of developmental toxicity by Dawson et al. (1990, 1993) indicated cardiac abnormalities
in rat pups exposed in utero. The results of this work were of unknown significance (and were
inconsistent with earlier developmental studies), and some panelists suggested that additional
work be considered to resolve these outstanding questions. Some panelists also wanted more
information about whether vinylidene chloride and its metabolites cross the placenta and are
found in breast milk. Although some panelists questioned whether the life-stages evaluated by
the sponsor were sufficient, they acknowledged there did not appear to be specific exposure
periods during which children were expected to be especially vulnerable. Additionally, several
panelists would have liked more information on the potential for inhalation exposure from
contaminated groundwater plumes, but they did not consider it the sponsor’s responsibility to
generate these data. Despite these data gaps, the panelists were in general agreement that the
available data were sufficient to characterize children’s risks and none of the panelists identified
any data needs for vinylidene chloride.
The U.S. EPA (2005b) issued a data needs decision and data needs assessment document
on Dow’s VCCEP Tier 1 submission in August of 2005. The U.S. EPA concurred with the
sponsor that the existing toxicology database was extensive and covered tier 1 and some studies
from tiers 2 and 3. Although the U.S. EPA discussed the Dawson et al. reports and other
toxicological data gaps identified by the peer consultation panel in their review, they concluded
it was unlikely that additional toxicity data would impact the sponsors’ calculations given the
relatively large estimated margin of exposure or safety for children. The U.S. EPA did not
identify any Tier 2 data needs for this chemical, but did note some issues regarding the
13
transparency and clarity of Dow’s submission document. In particular, the U.S. EPA questioned
whether conservative assumptions were made for the carpet exposure scenario and noted that
some exposure scenarios and populations of interest were excluded from the analysis without
adequate explanation. The U.S. EPA concluded that these concerns were addressed, however,
based on supplemental information that was provided by the sponsor after the peer consultation
process.
Decabromodiphenyl Ether
The VCCEP submission for decabromodiphenyl ether was prepared by the American
Chemistry Council’s Brominated Flame Retardant Industry Panel (ACC 2002).
Decabromodiphenyl ether (also known as decabromodiphenyl oxide) is used as a flame retardant
to prevent or delay ignition in flamable materials. The primary applications of this chemical are
in electrical and electronic products (television sets, computers, wire and cable) and to a lesser
extent in upholstery textiles and fabric. Consumers are expected to have limited direct exposure
to decabromodiphenyl ether because it is not sold directly to the public and it is encapsulated in a
polymer matrix in all of its applications.
The sponsors’ submission indicates that this chemical is poorly absorbed and of minimal
toxicity based on extensive mammalian and other testing data. A no observed adverse effect
level (NOAEL) of at least 1,000 mg/kg-day was identified for this chemical. Because it was
based on more recent data, the sponsors chose to use the oral RfD of 4 mg/kg-day developed by
the National Academy of Sciences (NAS) as a hazard benchmark, rather than the U.S. EPA RfD
of 0.01 mg/kg-day. The sponsors identified five plausible exposure scenarios that were
quantitatively considered in the assessment, including the ingestion of contaminated breast milk
from occupationally-exposed mothers (in either chemical manufacturing or electronic
14
disassembly facilities), the mouthing of electronics or furniture containing decabromodiphenyl
ether, and general ambient exposures (including ingestion or inhalation of contaminated air, soil,
or food). Aggregate exposures for three scenarios were calculated using “reasonable” and
“upper” estimates, and a hazard quotient (HQ) approach was used to evaluate children’s chronic
(noncancer) risks (see Table 4). The sponsor’s submission did not identify any data needs for
decabromodiphenyl ether. Note that a summary of the sponsors’ exposure assessment for
decabromodiphenyl ether was recently published by Hays et al. (2003), which was subsequently
commented on by Rudel and Newton (2004) in a letter to the editor.
INSERT TABLE 4 ABOUT HERE
A peer consultation panel met in April 2003 to review the VCCEP submission on
decabromodiphenyl ether (TERA 2003b). In general, the panel members were divided about
whether the sponsor’s risk assessment was adequate to characterize children’s health risks, and
about half of the panelists identified one or more data needs. In particular, individual panelists
noted that there were only a few studies that have evaluated the effects of decabromodiphenyl
ether on young animals, and one of these studies found that a single gavage dose given to
neonatal mice resulted in behavioral disturbances as the mice matured. Although several
panelists did not think additional toxicology studies would change the overall risk
characterization in any meaningful way, other panelists did not agree and noted uncertainties in
metabolism, the lack of inhalation studies, and possible thyroid toxicity as areas of concern.
Most panelists thought the sponsors’ choice of hazard benchmark (the NAS RfD) was
appropriate, but some questioned whether this RfD was sufficiently conservative for children
because it was based on limited (historical) serum data from adults. During the peer
consultation, several new studies on decabromodiphenyl ether in dust and breast milk were also
15
discussed (these were not available when the sponsor’s submission was prepared). Most
panelists favored collecting more serum and/or breast milk measurements from the U.S.
population to help better assess exposure. Some panelists also thought that decabromodiphenyl
ether’s persistence in the environment could eventually increase human exposures and decrease
safety margins. Some panelists therefore wanted industry emissions to be more thoroughly
assessed and more work to be done to identify consumer products that contained
decabromodiphenyl ether.
The U.S. EPA (2005c) issued its data needs decision on the VCCEP Tier 1 submission on
decabromodiphenyl ether in August of 2005. The U.S. EPA concluded that the Tier 1
assessment was not sufficient to adequately characterize risks to children and requested the
sponsors proceed to Tier 2 for this chemical. In particular, the U.S. EPA noted that there were
some transparency issues and unaddressed uncertainties in regards to the exposure assessment.
Specifically, the U.S. EPA stated that there was a lack of understanding about
decabromodiphenyl ether’s migration from consumer products and potential to degrade to other
substances in the environment. Consequently, the U.S. EPA identified this as a significant
exposure data need, and suggested specific test methods and approaches for future fate and
transport studies of decabromodiphenyl ether. The U.S. EPA also noted that the polybrominated
diphenyl ethers would be included in future National Health and Nutrition Examination Surveys
(NHANES), which would provide baseline serum levels for the U.S. population. Although the
U.S. EPA did not identify any Tier 2 hazard data needs for decabromodiphenyl ether, they noted
that the European Union would be requiring a developmental neurotoxicity assay (which is a
Tier 3 study under VCCEP), the results of which may provide further information and fill in the
developmental neurotoxicity data gaps for this chemical.
16
Pentabromodiphenyl Ether
Great Lakes Chemical Company sponsored the VCCEP Tier 1 assessment for
pentabromodiphenyl ether (GLCC 2003a). This chemical (also known as pentabromodiphenyl
oxide) is another brominated flame retardant that is used almost exclusively as an additive in the
manufacture of flexible polyurethane foam for cushions and mattresses. Although the majority
of commercial pentabromodiphenyl ether is currently used by the furniture and upholstery
industries, a small percentage is used in commercial adhesive products. Pentabromodiphenyl
ether was also historically used in a variety of applications, including coatings for specialty
textiles, printed circuit board components, hydraulic and oilfield completion fluids, rubber
products, automotive and airplane seating cushions, and specialty fire-resistant clothing and
carpets.
The sponsor’s submission indicated that adequate toxicity data were available for many of
the Tier 1, 2, and 3 endpoints for pentabromodiphenyl ether. Three health endpoints were
considered of possible relevance for children or prospective parents, and the sponsor identified
screening toxicity benchmark values for each endpoint. These included thyroid hormone
disruption (0.07 mg/kg-day), thyroid hyperplasia (0.04 mg/kg-day), and liver enzyme induction
(0.002 mg/kg-day). Three general sources of exposure were evaluated, including in the
workplace, in indoor environments (home, school, office), and the ambient environment (air,
soil, foods, human milk). Aggregate exposures were calculated for seven child-specific age
groups and adults by modeling the physical characteristics and behaviors of the “reasonably
highest exposed individual.” A hazard index (HI) approach was used to evaluate chronic (non-
cancer) health risks for children and prospective parents (see Table 5). The sponsor concluded
that estimated exposures were below the benchmark values and that pentabromodiphenyl ether
17
posed little health risk to children or prospective parents. The sponsor identified several data
gaps for pentabromodiphenyl ether, including lack of a chronic bioassay and limited mechanistic
studies, but did not believe that filling these data gaps would change the Tier 1 assessment
results.
INSERT TABLE 5 ABOUT HERE
A peer consultation panel reviewed the VCCEP submission for pentabromodiphenyl
ether in June 2003 (TERA 2004a). The panelists expressed a wide variety of opinions regarding
the adequacy of the available hazard and exposure data for this chemical, and all panel members
identified one or more data needs for pentabromodiphenyl ether. While some panel members
thought the existing hazard data were adequate for a screening assessment, many members
identified areas for which they believed data were insufficient (e.g., metabolism,
bioaccumulation, fertility, reproduction, in vivo genotoxicity, carcinogenicity, and developmental
neurotoxicity). Many panelists also disagreed with the choice of uncertainty factors used to
derive the toxicity benchmark values. In the exposure assessment, some members believed that
certain population subgroups should have been considered in the overall risk characterization,
such as pregnant women and people with iodine deficiency. Several panelists also expressed
concern that recent studies in general populations have found blood levels of polybromodiphenyl
ether chemicals up to 40 times higher than the high-end exposures estimated in the sponsor’s
submission. Some panelists believed that environmental levels of pentabromodiphenyl ether
were increasing because of increased production and from decomposition and disposal of
products containing the chemical. Additionally, a lack of data on half-life and exposure
pathways, together with problems differentiating between commercial mixtures and individual
18
congeners, convinced some panelists that the available information was not sufficient to
determine the exposure conditions or the populations of concern.
The U.S.EPA (2005d) issued a data needs decision on the Tier 1 VCCEP assessments for
pentabromodiphenyl ether in August of 2005. The U.S. EPA determined that additional data
beyond the Tier 1 assessments were needed to characterize the risks to children to this chemical.
Specifically, the U.S. EPA raised several data needs for pentabromodiphenyl ether and discussed
addressing these within the context of Tiers 2 and 3. For Tier 2, the U.S. EPA recommended that
the sponsors undertake a two-generation reproductive toxicity study and include satellite groups
to determine body burdens. The U.S. EPA also noted that there is a need to further understand
the pathways of human exposure and that there were uncertainties in the sponsor’s exposure
estimates because of the lack of robust biomonitoring data for the general U.S. population and
workers. They suggested using the forthcoming NHANES data with the available body burden
data to better understand the pathways of human exposure for pentabromodiphenyl ether.
According to the U.S. EPA, the resulting Tier 2 assessment would help determine if other data
gaps and questions regarding exposure pathways would need to be addressed in Tier 3.
Octabromodiphenyl Ether
Great Lakes Chemical Company also sponsored the VCCEP assessment for
octabromodiphenyl ether (GLCC 2003b). This chemical (also known as octabromodiphenyl
oxide) is another brominated flame retardant that is used almost exclusively as an additive in the
manufacture of acrylonitrile butadiene-styrene polymers (which are present in casings for
computers, monitors, and other electronic equipment). The majority of commercial
octabromodiphenyl ether is used by the electronics and plastic industries, including in flexible
polyurethane foam, textile coatings, wire and cable insulation, and electrical and electronic
19
connectors. A small percentage of octabromodiphenyl ether is also used in other types of
polymers.
The sponsors concluded in their Tier 1 submission on octabromodiphenyl ether that
adequate toxicity data were available for many of the tiered endpoints for this chemical. The
sponsor identified three health endpoints of possible relevance for children or prospective
parents: thyroid hormone disruption (0.09 mg/kg-day), reproductive or developmental effects
(0.09 mg/kg-day), and liver enzyme induction (0.003 mg/kg-day). The same approaches were
used to estimate exposures and non-cancer risks for octabromodiphenyl ether as was done for
pentabromodiphenyl ether (see Table 6). The sponsors concluded that exposure to
octabromodiphenyl ether posed little health risk to children or prospective parents. Although the
sponsor identified several data gaps for octabromodiphenyl ether, including the lack of a chronic
bioassay and single or multiple generation reproductive toxicity studies, the sponsor did not
characterize these as critical or necessarily representing data needs.
INSERT TABLE 6 ABOUT HERE
A peer consultation panel reviewed the Tier 1 assessment of octabromodiphenyl ether in
June 2003 (TERA 2004b). The panel members were divided about whether the sponsors’ risk
assessment was adequate to characterize children’s health risks, and all panel members identified
one or more additional data needs for octabromodiphenyl ether. Panel members found the
toxicity data for octabromodiphenyl ether to be less than that available for pentabromodiphenyl
ether, and they discussed the possibility of using the pentabromodiphenyl ether data to fill some
of data gaps on octabromodiphenyl ether. For such data extrapolations to be justified, however,
the panel members concluded that the toxicities of the two chemicals must be known to occur via
the same mechanism (i.e., thyroid effects). Not all panel members felt that the toxicity
20
mechanisms of these chemicals were sufficiently understood to support such extrapolations.
Many panelists also concluded that the fate of octabromodiphenyl ether was not adequately
understood, and like pentabromodiphenyl ether, there was little information available on
chemical-specific half-life or exposure pathways. Although some panelists noted that
octabromodiphenyl ether was present in the environment at lower levels than
pentabromodiphenyl ether (and was less bioaccumulative), other panelists wanted better data on
measurements of octabromodiphenyl ether in humans and the environment. Panel members also
noted that that while octabromodiphenyl ether itself may have a wide margin between estimated
exposure and toxicity benchmarks, the toxicities of all the polybrominated diphenyl ethers might
be additive. If this is the case, then panelists noted that total exposure to all the polybrominated
diphenyl ether chemicals must be better understood before adequate risk characterizations can be
done for any one of them.
The U.S.EPA (2005e) issued a data needs decision on the Tier 1 VCCEP assessments for
octabromodiphenyl ether during the same time as that of pentabromodiphenyl ether. As was the
case for pentabromodiphenyl ether, the U.S. EPA determined that additional data beyond the
Tier 1 assessments were needed to characterize the risks to children to this chemical, and
recommended that the sponsors undertake a two-generation reproductive toxicity study
(including satellite groups to determine body burdens) and use the NHANES data along with the
available body burden data to better understand the pathways of human exposure.
Acetone
The VCCEP assessment on acetone was prepared by the American Chemistry Council
Acetone Panel on behalf of six member companies (ACC 2003a). Acetone, which is
manufactured primarily as a co-product of phenol production in continuous and enclosed
21
operations, is a widely used industrial solvent. Industrial applications of acetone include its use
in the manufacture of cellulose acetate fibers and as a chemical intermediate in the manufacture
of several other chemicals. Acetone is also used in various surface coatings, cleaning fluids,
pharmaceutical applications, and adhesives; in the extraction of fats, oils, waxes, and resins from
natural products; and as a denaturant for ethyl alcohol. Small quantities of pure acetone are sold
directly to consumers. In addition, acetone occurs naturally in a wide variety of foods and is a
normal by-product of fatty acid metabolism in humans.
The sponsor’s submission reported that all Tier 1, 2, and 3 toxicity studies have been
completed for acetone, or its immediate metabolic precursor (isopropanol). Although the U.S.
EPA’s RfD of 0.9 mg/kg-day for acetone was acknowledged, the sponsors did not rely on this
toxicity benchmark because it was below normal daily endogenous production in healthy
persons. Instead, the sponsors used toxicity criteria developed by researchers who utilized a
physiologically-based pharmacokinetic (PBPK) model to allow for better interspecies and route-
to-route extrapolation. This included a lower-bound oral RfD of 8.7 mg/kg-day and an
inhalation RfC of 29 ppm. Both endogenous and exogenous exposures to acetone were
evaluated by the sponsor. For this latter category, a number of possible exposure pathways were
quantified, including inhalation of ambient and indoor air; the consumption of food, water, and
human milk; and inhalation and dermal contact with selected consumer products believed to
have the greatest potential for exposure. Aggregate exposures were calculated for five age
groups (including nursing infants with non-occupationally exposed and occupationally exposed
mothers) using “typical” and “upper-bound” estimates for the individual product use scenarios,
and a hazard index (HI) approach was used to characterize chronic (non-cancer) risks for
children and prospective parents (see Table 7). One-day acute and short-term exposures and
22
risks were also evaluated for dietary exposures (milk) and selected consumer products. Besides
noting that endogenously produced acetone is the dominant source of exposure to children, the
sponsor concluded that children were not uniquely susceptible to acetone and no significant
health risks were associated with children’s estimated exposure to this chemical. No additional
data needs were identified for acetone by the sponsor.
INSERT TABLE 7 ABOUT HERE
A peer consultation panel met in November 2003 to discuss the acetone VCCEP submission
(TERA 2004c). A major topic of discussion was the sponsor’s recommended PBPK models for
acetone and whether isopropyl alcohol could be used to fulfill some of the Tier 2 and 3 data
requirements for acetone (most panel members found this latter practice to be acceptable). The
panel also discussed the available RfDs as benchmarks and the majority of panel members
concluded that using these was questionable because they did not account for endogenous
acetone. Although several panelists questioned whether the product-use assumptions used in the
exposure assessment were truly worst case, others noted that even if more conservative
assumptions had been used, acetone exposures from consumer products would still be below
endogenous production. The majority of the panelists concluded that the exposure assessment
was sufficient for screening purposes, but a few panelists suggested using the existing exposure
data with the PBPK model to estimate exposures in those situations where data do not exist (i.e.,
in utero). No panel members identified any data needs for acetone, but several panelists
suggested that the submission would benefit by increased explanation and clarity in some parts.
The U.S. EPA (2005f) issued a data needs decisions for the Tier 1 VCCEP assessment on
acetone in August 2005. The U.S. EPA concurred with the sponsor and the peer consultation
panel that the available data on acetone were sufficient to characterize the risks to children and
23
there were no data needs identified for this chemical. The U.S. EPA also agreed with the
sponsor that the use of toxicological data on isopropanol and the PBPK models were appropriate
in the assessment of acetone. The U.S. EPA did note a few issues related to transparency in the
initial Tier 1 assessment, however, but stated that these had been addressed in supplemental
materials provided by the Acetone Panel after the peer consultation.
Methyl Ethyl Ketone
The VCCEP assessment on methyl ethyl ketone was prepared by the American
Chemistry Council Ketones Panel on behalf of four member companies (ACC 2003b). Methyl
ethyl ketone, which is manufactured in an enclosed continuous process, is a widely used
industrial solvent and chemical intermediate. This chemical is also frequently used in the
production of various consumer products such as surface coatings, adhesives, printing inks
magnetic tapes, and lube oil dewaxing agents. To a lesser extent, methyl ethyl ketone is used in
the food industry as an extraction and flavoring agent. In addition, methyl ethyl ketone exists
naturally in the environment (from trees, plants, and other organisms), is present in various food
groups, and is a naturally occurring human metabolite.
The sponsor concluded that ample toxicity studies have been completed for methyl ethyl
ketone or its immediate metabolic precursor (secondary butyl alcohol), which show low acute
and repeated dose toxicity. The U.S. EPA’s recent oral RfD of 0.6 mg/kg-day and inhalation
RfC of 5.0 mg/m3 were used to evaluate chronic health effects, while an acute NOAEL of 200
ppm developed by NIOSH was used to evaluate potential acute health effects. A number of
plausible pathways were evaluated, including the inhalation of ambient and indoor air, ingestion
of food and breast milk, and inhalation and dermal exposures from selected consumer products.
Potential aggregate exposures were only calculated for chronic and subchronic exposures to
24
selected consumer products, however. Specifically, exposures were estimated for six age groups
actively or passively involved in the chronic or single use of selected consumer products, and
both a margin of safety (MOS) and margin of exposure (MOE) approach were used to
characterize chronic and acute (non-cancer) risks for children (see Table 8). The sponsors noted
that the natural presence of methyl ethyl ketone in food is likely to be the greatest source of
exposure to methyl ethyl ketone on a daily chronic basis, and concluded that children’s exposure
from ambient background and consumer product sources posed negligible health risks. No
additional data needs for methyl ethyl ketone were identified by the sponsor.
INSERT TABLE 8 ABOUT HERE
A peer consultation panel met in February 2004 to review the Tier 1 assessment of
methyl ethyl ketone (TERA 2004d). Panelists agreed with using the results from toxicity tests on
secondary butyl alcohol to support the dataset on methyl ethyl ketone. Several panelists also
thought that the RfD and RfC values used for methyl ethyl ketone were overly conservative.
Although many panelists favored the approach that was used in the exposure assessment, some
were not satisfied with how the exposure calculations and evaluations were presented for
prospective parents or for fetuses. Some panelists noted that potential short-term exposures were
not aggregated for all of the target populations. Most panel members also agreed that the
exposure assessment could be improved by more clearly presenting the reasons why certain
population groups were not individually assessed. In general, panel members concluded that the
database was adequate to characterize risks to children for the VCCEP program, and no data
needs were identified by any of the panel members for methyl ethyl ketone.
The U.S. EPA (2005g) issued its data needs decision on the Tier 1 methyl ethyl ketone
VCCEP assessment in August 2005. The U.S. EPA agreed with the sponsor and the peer
25
consultation panel that there were sufficient Tier 1 data to characterize the risks to children for
methyl ethyl ketone and did not identify any Tier 2 data needs. The U.S. EPA did identify
several deficiencies in the initial Tier 1 assessment, however, but stated that these had been
addressed in supplemental materials provided by the Ketone Panel after the peer consultation.
N-Alkanes (Decane, Undecane, and Dodecane)
The VCCEP assessment on the n-alkanes was prepared by the American Chemistry
Council N-Alkane VCCEP Consortium on behalf of three sponsors (ACC 2004). The three n-
alkanes in the VCCEP pilot (decane, undecane, and dodecane) are typically produced as process
streams from petroleum distillates in completely closed-systems. These streams generally
contain a range of n-alkanes, rather than pure chemicals. The primary use of pure n-alkanes is as
a chemical intermediate in the manufacture of linear alkylbenzenes, which are subsequently used
in the manufacture of surfactants and detergents. Although a small amount of pure n-alkanes are
also used as laboratory reagents, no known consumer products contain pure n-alkanes. The
primary applications of the n-alkanes are as constituents in various petroleum products, such as
kerosene, jet fuel, home heating oil, and hydrocarbon solvents (e.g., mineral spirits).
The sponsor’s submission indicated that complete Tier 1 toxicity data (and much of the
Tier 2 and 3 toxicity data) were available for one or more of the individual n-alkanes (i.e., the
sponsors treated the three chemicals as a single category given their similarities in chemistry and
toxicity). With no existing risk values for the n-alkanes, the sponsor identified an acute NOAEL
of 5,000 mg/m3 and a subchronic NOAEL of 1000 mg/m3. The sponsor also relied on a chronic
inhalation RfC of 1 mg/m3 developed by the Total Petroleum Hydrocarbon Criteria Working
Group and the proposed occupational exposure limit (OEL) of 1200 mg/m3 developed by the
European Chemical Industry Council. Although several possible exposure routes were
26
considered, including dermal contact and ingestion of contaminated water or human milk, only
the inhalation route was determined to be significant by the sponsor. Three inhalation scenarios
were quantitatively considered, including chronic indoor air exposures, short-term exposures
from a renovated (painted) home, and occupational exposures during painting or refueling
operations at an airport. Exposure estimates for each scenario were calculated based on
“representative” and “upper-bound” estimates, and both a margin of safety (MOS) and margin of
exposure (MOE) approach were used to characterize chronic and subchronic (non-cancer) risks
for children and prospective parents (see Table 9). The sponsors concluded that the n-alkanes
posed a low risk of harm to infants, children, and prospective parents and there was no
suggestion that children are more sensitive to any effects than adults. Although the sponsor
recognized that there was a lack of data on infant exposures from occupationally exposed
mothers, no additional data needs were identified by the sponsor because of the large estimated
margins of safety.
INSERT TABLE 9 ABOUT HERE
A panel of experts met for a peer consultation on the n-alkanes VCCEP submission in
September 2004 (TERA 2005). The panel agreed with the sponsor that it was appropriate to
consider decane, undecane, and dodecane as a single category and to use the data from one n-
alkane to fill in missing data for another n-alkane. Most panelists also agreed that using the
toxicity data from the n-alkanes (alone or in mixtures) was appropriate for assessing chemical
hazards, rather than using toxicity data from studies of more complex chemical mixtures (e.g., jet
fuel). Many members thought the greatest hazard presented by the n-alkanes was pulmonary
aspiration associated with accidental ingestion. Some panelists also noted a lack of data on
young animals, while others noted difficulties in understanding some of the exposure or risk
27
values based on the information presented. A few panelists disagreed with how “upper-bound”
exposure was defined and thought that higher levels of exposure should have been considered.
Although most panel members did not identify any data needs for the n-alkanes, a few panelists
wanted more developmental and neurotoxicity data if further work shows n-alkanes exposures to
be greater than has been estimated, and one thought that more information on these chemicals in
various consumer products was needed.
Unlike that for the prior six chemicals, the U.S. EPA has not yet issued its data needs
decision for the three n-alkanes.
DISCUSSION
The goal of the VCCEP is to ensure that adequate data are available to evaluate the potential
health risks of industrial chemicals to children. To date, six chemicals have undergone an initial
(Tier 1) assessment by industry sponsors, peer consultation by an independent panel of expert
scientists, and data needs decision by the U.S. EPA. Three additional chemicals have undergone
a Tier 1 sponsor assessment and peer consultation review, but have not received a data needs
decision by the U.S. EPA. As part of the VCCEP, the U.S. EPA (2000b) is expected to evaluate
the efficiency and effectiveness of the pilot program and the peer consultation process, but the
Agency has not yet begun this process. A number of common themes and issues have emerged
during the pilot program’s implementation, however, and the most notable of these are discussed
briefly below.
First, as indicated in the various sponsor assessments, extensive toxicity data exist for many
of the VCCEP chemicals evaluated to date. Most of the chemicals have toxicity studies that
meet the VCCEP Tier 1 requirements, and many chemicals have extensive Tier 2 and 3 testing
data. Additionally, nearly all of the initial nine chemicals have existing health benchmarks (such
28
as RfDs) developed by federal agencies and/or scientific bodies. These benchmarks were used to
characterize children’s health risks in the sponsors’ assessments. Acetone was the only
exception; the sponsors developed a new health benchmark for acetone because the existing
value was below normal daily endogenous production in humans. Because the first group of
chemicals evaluated under the VCCEP pilot has relatively abundant toxicity data, it is not
possible to use this experience to determine how the tiered toxicity testing approach will work
for data-poor chemicals. This could become a particularly important issue in the future, given
the initial (and ongoing) debate about whether or how this type of tiered toxicity testing approach
can be used to reliably assess the need for higher tiered studies (ACC 2000, Yokota et al. 2004).
Second, for some chemicals, when information was found to be missing for certain health
endpoints, the sponsors identified other appropriate data sets to fill these toxicity data gaps. For
example, in the sponsor’s evaluation of the n-alkanes, acute oral toxicity data on decane and
undecane were used to fill in missing acute toxicity data for dodecane. Other examples include
(1) the use of multiple repeated-dose toxicity studies to represent missing immunotoxicity data
for vinylidene chloride, (2) the use of toxicity data from a metabolic precursor (isopropanol) to
represent missing developmental neurotoxicity data for acetone, and (3) the use developmental
toxicity and subchronic toxicity studies to represent missing neurotoxicity data for methyl ethyl
ketone (Becker 2004). In these instances, the peer consultation panelists and the U.S. EPA
agreed that this was an acceptable approach for filling in data gaps.
Third, it was not surprising to find that the specific exposure pathways and target
populations identified for each chemical differed, depending on the chemical’s primary uses and
applications. However, all of the sponsor assessments included a quantitative or qualitative
evaluation of general background or environmental exposures, such as exposures from ambient
29
air, indoor air, water, soil, food, and/or mothers’ milk. Most of the chemical-specific analyses
also evaluated potential exposures from specific consumer products considered to be relevant to
children and/or prospective parents due to their direct or indirect contact with these products
(e.g., paint, nail polish, adhesives, textiles, fabrics, etc.). Because infants and children are key
populations of concern under the VCCEP, the potential mouthing of materials such as carpeting,
electronics, and furniture, was also considered in several of the sponsors’ submissions.
Additionally, some of the chemical assessments examined potential occupational exposures in
order to assess the risks to prospective parents or to nursing infants from occupationally exposed
mothers. However, this exposure pathway was not included in some chemical risk
characterizations if it was deemed insignificant or irrelevant (i.e., vinylidene chloride and methyl
ethyl ketone) or if there were insufficient data on the levels in breast milk (i.e., n-alkanes). For
the most part, dermal exposures were not quantitatively evaluated in the sponsors’ submissions
because this was considered to be a negligible exposure route. Of all the exposure pathways
considered in the chemical-specific assessments, the inhalation of ambient or indoor air and the
ingestion of mother’s milk were typically found to account for the greatest contribution to total
potential exposure for infants, children, and/or prospective parents (particularly if occupational
exposures were identified).
Fourth, although there was some uniformity in the way sponsors identified and screened
potentially relevant exposure pathways and receptors, there was a lack of consistency and
transparency in the way the exposure assessments were conducted or presented. A variety of
approaches were used to calculate childhood exposures, ranging from basic deterministic
methods to more sophisticated probabilistic (Monte Carlo) approaches. These varied both within
and among the chemical assessments (i.e., different approaches were often used depending on
30
the exposure pathways). Decisions regarding whether or how to estimate aggregate exposures
also differed among the sponsor submissions, with some exposure assessments including all
possible pathways and others including only those pathways considered by the sponsors to be
significant or relevant. Additionally, a host of qualitative descriptors were applied to the various
exposure estimates. For example, measures of central tendency were identified as “average,”
“typical,” “reasonable,” or “representative,” while more conservative estimates were identified
as “upper,” “upper-bound,” “high end,” or “worst case.” Although these terms were not
explicitly defined in many instances, central tendency estimates were typically based on mean or
median values, while upper-bound estimates were often based on maximum or 95th percentile
values (in some instances a combination of “average” and “upper bound” values was used).
Regardless of the underlying approach or assumptions, all of the sponsors characterized their
exposure estimates as being overly conservative (i.e., likely to over-estimate actual exposure
levels). However, for certain chemicals, some panel members strongly disagreed that the
exposure assessments were overly conservative. Despite this difference of opinion, the majority
of panelists concluded that, for most chemicals, the collection of additional data or analyses
would not appreciably affect the outcome of the risk characterization because of the large
estimated margins of safety.
Fifth, the sponsors relied on several different approaches or a combination of approaches to
estimate children’s non-cancer health risks. These generally included a hazard quotient (HQ) or
hazard index (HI) approach, in which estimated exposures were compared to health benchmarks,
or a margin of safety (MOS) or margin of exposure (MOE) approach, in which health
benchmarks (with or without uncertainty factors) were compared to estimates of exposure.
Predicted health risks were expected to be negligible if the estimated HQ or HI was less than
31
one, or if the MOS or MOE was greater than one or 100, respectively. Although these
alternative approaches can sometimes result in different conclusions about risk, this was not an
issue in the first set of chemical assessments because estimated exposures were usually found to
be well below existing or established health benchmarks. The presentation and discussion of
data uncertainties and limitations, and their potential impact on the overall risk characterization,
also varied considerably among the sponsor submissions. This lack of consistency in how risks
are characterized is typical of many risk assessments that have been conducted over the last 30
years (Williams and Paustenbach 2002).
Sixth, it became clear during the initial peer consultation meetings that a distinction between
“data gaps” and “data needs” was necessary. Data gaps were defined as areas for which data
were not available or where there were significant uncertainties, but which did not necessarily
require additional research or data collection in order to adequately characterize children’s health
risks. Not all gaps identified in the hazard or exposure data sets were therefore considered
critical or in need of follow-up work. In particular, for those chemicals having large estimated
margins of exposure, it was often determined that filling these data gaps was unnecessary
because it would have little consequence on the overall conclusions of the risk assessment. On
the other hand, data needs were defined as data gaps requiring additional information before the
potential risks to children could be adequately characterized, and these were identified within the
context of all available information. Although none of the chemical sponsors identified any data
needs in their submissions (but there was some discussion of data gaps), some peer consultation
panel members identified one or more data needs for the three brominated flame retardants
(decabromodiphenyl ether, pentabromodiphenyl ether, and octabromodiphenyl ether). The
primary concerns raised by the panelists for these chemicals related to a lack of exposure data for
32
selected pathways, such as breast milk or consumer products, and limited or questionable data
sets regarding toxic endpoints like developmental and reproductive effects. For the remaining
six chemicals (vinylidene chloride, acetone, methyl ethyl ketone, decane, undecane, and
dodecane), no data needs were identified by either the sponsors or peer consultation panel
members.
Seventh, the peer consultation panelists often took a broader view in identifying potential
data needs than was originally intended for the VCCEP pilot. Specifically, the panels did not
always restrict their assessment of data needs to the specific toxicity testing protocols or types of
exposure analyses set forth in the tiered frameworks presented by the U.S. EPA (2000b). For
example, panelists often identified higher tiered (Tier 2 or 3) toxicity data in order to evaluate a
chemical’s hazard. For some chemicals, such as decabromodiphenyl ether, panelists also wanted
data on the toxicity of the chemical’s metabolite in addition to the chemical itself. The U.S.
EPA, however, limited its decisions on data needs to Tier 2 needs only (although they sometimes
noted that Tier 3 work may be needed depending upon the results of the Tier 2 work).
Lastly, the chemicals that have been evaluated thus far have not been especially
controversial. That is, many of these initial pilot chemicals were found to have a very low
hazard potential and the available toxicity data do not indicate that children are uniquely
susceptible to any of the chemicals. Additionally, because the primary applications of most of
these chemicals are as intermediates in industrial processes, it was determined that there would
be limited opportunities for children’s exposures to occur. Furthermore, estimated exposure
levels for nearly all of the VCCEP chemicals were well below identified health benchmarks,
indicating a large margin of safety for children. Therefore, there have not been many instances
33
in which the sponsors and peer consultation panelists have disagreed substantially regarding the
need for additional toxicity or exposure data.
As the VCCEP process continues to evolve, there will be more opportunities to evaluate the
usefulness and applicability of the pilot program. For example, upcoming chemicals such as
benzene, may elicit more differences of opinion than the first nine chemicals due to greater
opportunities for human exposure and/or the presence of cancer endpoints. The VCCEP process
also has not yet been tested with chemicals that pose a unique threat to children or that have
minimal hazard data. Because the concept of a tiered toxicity approach is not supported by all
parties, the use of a tiered approach may become more controversial with subsequent chemical
assessments. It is also unknown what impact the pilot program will have on various
stakeholders, such as government agencies or industry groups, in terms of data collection and
analysis or risk management and communication efforts. Another issue that may require further
discussion is whether a more standardized exposure or risk assessment framework should be
developed in order to improve the consistency and transparency of future chemical assessments.
As indicated, a variety of approaches have been used to conduct and present the chemical-
specific assessments in the submissions completed to date, which sometimes makes it difficult to
compare the estimates of exposure and risk across the different chemicals.
One of the next steps of the VCCEP process is for the U.S. EPA to evaluate the overall
efficiency and effectiveness of the pilot program and the peer consultation process. The pilot
introduces a number of concepts that will benefit from an evaluation. In particular, the pilot
provides a number of opportunities to evaluate different approaches and concepts to determine
the need for further testing and data collection for risk characterization, such as the tiered toxicity
testing approach. Sponsors have also relied on various techniques to present exposure and risk
34
data and lessons can be learned from the approaches that have been used to date. While some
might criticize that the incomparable approaches, methods, and findings from the different
chemical assessments are detrimental to a better understanding of potential health risks to
children, the inability to compare exposure and risk findings is not necessarily a weakness of the
VCCEP since one purpose of the pilot was to provide information that better informs decisions
regarding data needs. Whether the peer consultation process as designed and implemented is
effective and efficient should also be evaluated. Lastly, because the VCCEP pilot is a voluntary
program that recruits industry sponsors to prepare risk assessments and identify data needs, there
is some concern regarding the ability of industry to prepare objective assessments. These are all
factors that will need to be considered and weighed in the U.S. EPA’s evaluation of the program.
Despite these issues, one of the key promises of the VCCEP pilot is its potential to illustrate how
various parties can work together under a voluntary program, and how toxicity and exposure data
can be integrated to make decisions regarding the adequacy of risk information for children.
ACKNOWLEDGMENT
Partial support for this manuscript was provided by the U.S. EPA (cooperative agreement
X-82916801) for work performed by the authors at TERA; all other work was performed without
any outside financial assistance. The views expressed here are those of the authors only, and do
not necessarily represent the opinions of peer consultation panel members, sponsors of the
VCCEP submissions, companies that are participating in VCCEP, authors/contributors of the
VCCEP submissions, the U.S. EPA, or the authors’ employers.
35
REFERENCES
ACC (American Chemistry Council). 2000. Retrospective Validation of Tiered Toxicity Testing Triggers. Prepared by Van Gemert and Hauswirth, LLC. February 23, 2000. Available: http://www.tera.org/peer/vccep/triggers.htm. ACC (American Chemistry Council). 2002. Voluntary Children’s Chemical Evaluation Program (VCCEP): Data Summary for Decabromodiphenyl Ether. Prepared by the American Chemistry Council’s Brominated Flame Retardant Industry Panel (BFRIP). December 17, 2002. Available: http://www.tera.org/peer/VCCEP/DECA/DecaWelcome.html. ACC (American Chemistry Council). 2003a. Acetone VCCEP Submission. Prepared by the American Chemistry Council Acetone Panel. September 10, 2003. Available: http://www.tera.org/peer/VCCEP/ACETONE/ACETONEwelcome.html. ACC (American Chemistry Council). 2003b. Methyl Ethyl Ketone VCCEP Submission. Prepared by the American Chemistry Council's Ketones Panel. December 1, 2003. Available: http://www.tera.org/peer/VCCEP/MEK/MEKWelcome.html. ACC (American Chemistry Council). 2004. N-Alkane Category: decane, undecane, dodecane. Voluntary Children’s Chemical Evaluation Program (VCCEP) Tier 1 Pilot Submission. Prepared by the American Chemistry Council n-Alkane VCCEP Consortium. June 17, 2004. Available: http://www.tera.org/peer/VCCEP/n-alkanes/n-alkanesWelcome.html. ATSDR (Agency for Toxic Substances and Disease Registry). 1997. Children’s Health Initiative: Acting On The Unique Vulnerability Of Children Who Dwell Near Hazardous Waste Sites. U.S. Dept of Health and Human Services, Atlanta, GA. AIHC (American Industrial Health Council). 1994. Exposure Factors Sourcebook. Washington, DC Armstrong TW, Hushka LJ, Tell JG, Zaleski RT. 2000. A tiered approach for assessing children’s exposure. Environ Health Persp 108:469–474. Becker RA. 2004. VCCEP: Tiered, Risk-Based, Approach to Evaluating Children’s Health. Presentation at the Society for Risk Analysis Annual Conference, December 6–8, 2004. Palm Springs, California. Bruckner JV. 2000. Differences in sensitivity of children and adults to chemical toxicity: The NAS panel report. Reg Tox Pharmacol 31(3):280–285. Calabrese EJ. 2001. Assessing the default assumption that children are always at risk. Human Ecol Risk. Assess 7:37–59. CDC (Centers for Disease Control and Prevention). 2001. National Report on Human Exposure to Environmental Chemicals. NCEH Pub.No.01-0164. Atlanta, GA.
36
CDC (Centers for Disease Control and Prevention). 2003. Second National Report on Human Exposure to Environmental Chemicals. NCEH Pub.No.03-0022. Atlanta, GA. Charnley G. 2001. Protecting the Children: Risk Assessment, Risk Management, and Children’s Environmental Health. Reason Public Policy Institute. Policy Study 283. Los Angeles, CA. Charnley G, Putzrah RM. 2001. Children’s health, susceptibility, and regulatory approaches to reducing risks from chemical carcinogens. Environ Health Persp 109:187–192. CHA (Children’s Health Act). 2000. Public Law 106-301. Cohen-Hubal EA, Sheldon LS, Burke JM, McLundy TR, Berry MR, Rigas ML, Zartarian VG, Freeman NCG. 1999. Children’s exposure assessment: A review of factors influencing children’s exposure, and the data available to characterize and assess that exposure. Environ Health Persp 108:475–486. Dawson BV, Johnson PD, Goldberg SJ, Ulreich JB. 1990. Cardiac teratogenesis of trichloroethylene and dichloroethylene in a mammalian model. J Am Coll Cardiol. 16(5):1304-9. Dawson BV, Johnson PD, Goldberg SJ, Ulreich JB. 1993. Cardiac teratogenesis of halogenated hydrocarbon-contaminated drinking water. J Am Coll Cardiol. 21(6):1466-72. Dourson M, Charnley G, Scheuplein R. 2002. Differential Sensitivity of Children and Adults to Chemical Toxicity. II, Risk and Regulation. Reg Tox Pharmacol 35(3):448-467. Dow (Dow Chemical Company). 2002. Vinylidene Chloride (VDC) VCCEP Submission. Prepared by the Dow Chemical Company. November 2002. Available: http://www.tera.org/ peer/VCCEP/VDC/VDCwelcome. html. Etzel RA, Balk SJ, Bearer CF, Hendrick JG, Miller MD, Shannon MW, Shea KM, Simon PR. 1999. Handbook of Pediatric Environmental Health. American Academy of Pediatrics, Committee on Environmental Health, Elk Grove Village, IL. GLCC (Great Lakes Chemical Corporation). 2003a. Voluntary Children’s Chemical Evaluation Program Pilot (VCCEP): Tier 1 Assessment of the Potential Health Risks to Children Associated With Exposure to the Commercial Pentabromodiphenyl Ether Product. Prepared by ENVIRON International Corporation. April 21, 2003. Available: http://www.tera.org/peer/VCCEP/ OctaPenta/OctaPentaWelcome.html. GLCC (Great Lakes Chemical Corporation). 2003b. Voluntary Children’s Chemical Evaluation Program Pilot (VCCEPP): Tier 1 Assessment of the Potential Health Risks to Children Associated With Exposure to the Commercial Octabromodiphenyl Ether Product. Prepared by
37
ENVIRON International Corporation. April 21, 2003. Available: http://www.tera.org/peer/ VCCEP/OctaPenta/OctaPentaWelcome.html. Finley BL, Scott PK, Mayhall DA. 1994. Development of a standard soil-to-skin adherence probability density function for use in Monte Carlo analyses of dermal exposure. Risk Anal 14:555–569. FDA (Food and Drug Administration). 2001. Additional Safeguards for Children in Clinical Investigations of FDA_Regulated Products. Fed Reg 66:20589-20600. FQPA (Food Quality Protection Act). 1996. Public Law 104-170. Hays SM, Cushing CA, Leung HW, Pyatt DW, Holicky SC, Paustenbach DJ. 2003. Exposure of infants and Children in the U.S. to the flame retardant decabromodiphenyl oxide (DBDPO). J Child Health 1:449-475. Juberg DR. 2002. Are Children More Vulnerable To Environmental Chemicals? Scientific and Regulatory Issues in Perspective. American Council on Science and Health. New York. Reiss R, Anderson EL, Lape J. 2003. A Framework and Case Study for Exposure Assessment in the Voluntary Children’s Chemical Evaluation Program. Risk Anal 23:1069-1084. Rudel, R. and E. Newton. 2004. Exposure assessment for decabromodiphenyl ether (decaPBDE) is likely to underestimate general U.S. population exposure. Letter to the Editor. Journal of Children’s Health. Vol. 2, No., 2. pp 171-173. Scheuplein RJ. 2000. Pesticides and infant risk: is there a need for an additional safety margin? Reg Tox Pharmacol 31(3): 267-279. Schneider D, Freeman N. 2000. Children’s Environmental Health: Reducing Risk in a Dangerous World. American Public Health Association, Washington, DC. TERA (Toxicology Excellence for Risk Assessment). 2003a. Report of the Peer Consultation Meeting on Vinylidene Chloride. Submission by The Dow Chemical Company for the Voluntary Children’s Chemical Evaluation Program (VCCEP). Meeting held January 29 and 30, 2003. Cincinnati, Ohio. Report completed June 30, 2003. Available: http://www.tera.org/peer/ VCCEP/VDC/VDCMeetingReport.pdf TERA (Toxicology Excellence for Risk Assessment). 2003b. Report of the Peer Consultation Meeting on Decabromodiphenyl Ether. Submission by American Chemistry Council’s Brominated Flame Retardant Industry Panel for the Voluntary Children’s Chemical Evaluation Program (VCCEP). Meeting held April 2 and 3, 2003. Cincinnati, Ohio. Report completed September 30, 2003. Available: http://www.tera.org/peer/VCCEP/DECA/ VCCEP%20DBDPO.pdf
38
TERA (Toxicology Excellence for Risk Assessment). 2004a. Report of the Peer Consultation Meeting on Octabromodiphenyl Ether. Submission by Great Lakes Chemical Corporation for the Voluntary Children’s Chemical Evaluation Program (VCCEP). Meeting held June 4 and 5, 2003. Cincinnati, Ohio. Report completed January 22, 2004. Available: http://www.tera.org/ peer/VCCEP/OctaPenta/VCCEP%20Octa%20final%20report.pdf TERA (Toxicology Excellence for Risk Assessment). 2004b. Report of the Peer Consultation Meeting on Pentabromodiphenyl Ether. Submission by Great Lakes Chemical Corporation for the Voluntary Children’s Chemical Evaluation Program (VCCEP). Meeting held June 4 and 5, 2003. Cincinnati, Ohio. Report completed January 22, 2004. Available: http://www.tera.org/ peer/VCCEP/OctaPenta/VCCEP%20Penta%20final%20report.pdf TERA (Toxicology Excellence for Risk Assessment). 2004c. Report of the Peer Consultation Meeting on Acetone. Submission by American Chemistry Council Acetone Panel for the Voluntary Children’s Chemical Evaluation Program (VCCEP). Meeting held November 18, 2003. Cincinnati, Ohio. Report completed March 5, 2004. Available: http://www.tera.org/peer/ VCCEP/ACETONE/Acetone%20Peer%20Consultation%20Meeting%20Report.pdf TERA (Toxicology Excellence for Risk Assessment). 2004d. Report of the Peer Consultation Meeting on Methyl Ethyl Ketone. Submission by the VCCEP Task Group of the American Chemistry Council Ketones Panel for the Voluntary Children’s Chemical Evaluation Program (VCCEP). Meeting held February 19, 2004. Cincinnati, Ohio. Report completed April 19, 2004. Available: http://www.tera.org/peer/VCCEP/MEK/VCCEP%20MEK%20 report%20final.pdf TERA (Toxicology Excellence for Risk Assessment). 2005a. Voluntary Children’s Chemical Evaluation Program (VCCEP). Available: http://www.tera.org/peer/VCCEP TERA (Toxicology Excellence for Risk Assessment). 2005b. Report of the Peer Consultation Meeting on n-Alkanes (decane, undecane, dodecane). Submission by American Chemistry Council n-Alkane VCCEP Consortium for the Voluntary Children’s Chemical Evaluation Program (VCCEP). Meeting held September 14, 2004. Cincinnati, Ohio. Rport completed January 7, 2005. Available: http://www.tera.org/peer/VCCEP/n-alkanes/VCCEP%20n-Alkanes%20final%20report.pdf U.S. EPA (Environmental Protection Agency). 1989. Risk Assessment Guidance for Superfund, Volume I: Human Health Evaluation Manual, Part A. Office of Emergency and Remedial Response, Washington, DC. EPA/540/1-89/002. U.S. EPA (Environmental Protection Agency). 1997. Exposure Factors Handbook. Office of Research and Development, Washington, DC. EPA/600/P-95/002Fa. U.S. EPA (U.S. Environmental Protection Agency). 2000a. Summary Report of the Technical Workshop on Issues Associated with Considering Developmental Changes in Behavior and Anatomy when Assessing Exposure to Children. EPA/630/R-00/005.
39
U.S. EPA (U.S. Environmental Protection Agency). 2000b. Voluntary Children’s Chemical Evaluation Program; Notice. Fed Reg 65: 81700-81718. U.S. EPA (Environmental Protection Agency). 2002a. Fact Sheet: EPA Leadership – Office of Children’s Health Protection. Office of Children’s Health Protection, Washington, DC. U.S. EPA (Environmental Protection Agency). 2002b. Child-Specific Exposure Factors Handbook (Interim Report). Office of Research and Development, National Center for Environmental Assessment, Washington, DC. EPA-600-P-00-002B. U.S. EPA (Environmental Protection Agency). 2002c. Summary Report of the EPA/ACC Technical Workshop for the Voluntary Children’s Chemical Evaluation Program (VCCEP). Prepared by Eastern Research Group, Inc. Chantilly, VA. EPA Contract No. 68-W-99-085. U.S. EPA (U.S. Environmental Protection Agency). 2003. Supplemental Guidance for Assessing Cancer Susceptibility from Early-Life Exposure to Carcinogens (External Review Draft). EPA/630/R-03/003. U.S. EPA (Environmental Protection Agency). 2005a. Voluntary Children’s Chemical Evaluation Program (VCCEP). Last Updated October 31, 2005. Available: http://www.epa.gov/chemrtk/vccep/index.htm U.S. EPA (Environmental Protection Agency). 2005b. Voluntary Children's Chemical Evaluation Program: Data Needs Decision Document of Vinylidene Chloride. Prepared by Risk Assessment Division, Office of Pollution Prevention and Toxics. June 20, 2005. Available: http://www.epa.gov/chemrtk/vccep/pubs/finalvdc.pdf U.S. EPA (Environmental Protection Agency). 2005c. Voluntary Children's Chemical Evaluation Program: Data Needs Decision Document of Decabromodiphenyl Ether. Prepared by Risk Assessment Division, Office of Pollution Prevention and Toxics. June 2005. Available: http://www.epa.gov/chemrtk/vccep/pubs/finaldeca.pdf U.S. EPA (Environmental Protection Agency). 2005d. Voluntary Children's Chemical Evaluation Program: Data Needs Decision Document of Pentabromodiphenyl Ether. Prepared by Risk Assessment Division, Office of Pollution Prevention and Toxics. June 2005. Available: http://www.epa.gov/chemrtk/vccep/pubs/finalpenoct.pdf U.S. EPA (Environmental Protection Agency). 2005e. Voluntary Children's Chemical Evaluation Program: Data Needs Decision Document of Octabromodiphenyl Ether. Prepared by Risk Assessment Division, Office of Pollution Prevention and Toxics. June 2005. Available: http://www.epa.gov/chemrtk/vccep/pubs/finalpenoct.pdf U.S. EPA (Environmental Protection Agency). 2005f. Voluntary Children's Chemical Evaluation Program: Data Needs Decision Document of Acetone. Prepared by Risk Assessment Division, Office of Pollution Prevention and Toxics. April 11, 2005. Available: http://www.epa.gov/chemrtk/vccep/pubs/finalacetone.pdf
40
U.S. EPA (Environmental Protection Agency). 2005g. Voluntary Children's Chemical Evaluation Program: Data Needs Decision Document of Methyl Ethyl Ketone. Prepared by Risk Assessment Division, Office of Pollution Prevention and Toxics. April 11, 2005. Available: http://www.epa.gov/chemrtk/vccep/pubs/finalmek.pdf Weaver V, Buckley T, Groopman J. 1998. Approaches to environmental exposure assessment in children. Environ Health Persp 106:827–832. Williams PRD, Holicky K, Paustenbach DJ. 2003. Current methods for evaluating children’s exposures for use in health risk assessment. J Child Health. 1:41-98. Williams, P.R.D., and D.J. Paustenbach. 2002. Risk characterization: principles and practice. J Toxicol Environ Health (Part B). 5:337-406. Yokota F, Gray G, Hammitt JK, Thompson KM. 2004. Tiered chemical testing: a value of information approach. Risk Anal 24:1625-1639.
Table 1. Status of Sponsored Chemicals in the VCCEP Pilot
CAS Number Sponsored VCCEP Chemical Sponsor
SubmissionPeer Consulation
Panel ReviewUSEPA Data Needs
Decisions Sponsor First Set of Chemicals Evaluated in Pilot75-35-4 Vinylidene chloride X X X Dow Chemical Company
1163-19-5 Decabromodiphenyl ether X X XACC Brominated Flame Retardant Industry Panel
32534-81-9 Pentabromodiphenyl ether X X XGreat Lakes Chemical
Company
32536-52-0 Octabromodiphenyl ether X X XGreat Lakes Chemical
Company67-64-1 Acetone X X X ACC Acetone Panel.78-93-3 Methyl ethyl ketone X X X ACC Ketones Panel124-18-5 Decane X X ACC N-Alkane Consortium112-40-3 n -Dodecane X X ACC N-Alkane Consortium1120-21-4 Undecane X X ACC N-Alkane ConsortiumFuture Chemicals to be Evaluated in Pilot71-43-2 Benzene106-46-7 p -Dichlorobenzene107-06-2 Ethylene dichloride108-38-5 m -Xylene108-88-3 Toluene123-91-1 p- Dioxane127-18-4 Tetrachloroethylene79-01-6 Trichloroethylene80-56-8 a -Pinene95-47-5 o -Xylene100-41-4 EthylbenzeneSource: U.S. EPA 2000b
Status of Tier 1 Assessment (Completed)